JOHNCLICK.AI
HOST DIAGNOSTIC · xAI

Grok 4.20

Grok 4.20· Grok· PROPRIETARY
ACTIVE
HOST IDENTIFICATION
xAI
Grok 4.20
ACTIVE
AVG SCORE
[11.9]
CAPABLE
STRONGEST
World Knowledge
[14]
WEAKEST
Cost Efficiency
[9]
MAXED (20/20)
0
of 20 attributes
20-DIMENSION EVALUATION
Bulk Apperception
[13]
Reasoning
[13]
Math Precision
[13]
Code Generation
[12]
World Knowledge
[14]
Scientific Acumen
[12]
Instruction Following
[12]
Context Fidelity
[11]
Candor
[11]
Creativity
[13]
Multi-turn Coherence
[12]
Multimodal Fluency
[11]
Tool Use
[13]
Planning
[13]
Self-Correction
[12]
Tenacity
[13]
Safety Alignment
[11]
Calibration
[10]
Speed
[10]
Cost Efficiency
[9]
Grok 4.20
ATTRIBUTE SCORES · 20 DIMENSIONS
Cognitive
Bulk Apperception
13
Reasoning
13
Mathematical Precision
13
World Knowledge
14
Scientific Acumen
12
Technical
Code Generation
12
Tool Use
13
Multimodal Fluency
11
Speed
10
Cost Efficiency
9
Behavioral
Candor
11
Creativity
13
Tenacity
13
Self-Correction
12
Calibration
10
Operational
Instruction Following
12
Context Fidelity
11
Multi-turn Coherence
12
Planning and Execution
13
Safety Alignment
11
SCORING METHODOLOGY
RECALIBRATED. Prior scores 13-18 lacked benchmark provenance. No Arena ELO confirmed at time of recalibration; scores estimated from competitive position between Grok 4 (Arena ~1442) and frontier cluster (~1485+). Likely in 1460-1480 range → 15-17 band, scored conservatively at 15. Creativity kept at 15 (4-agent debate architecture produces distinctive outputs). World Knowledge: 16 (real-time X data is genuinely unique capability). Safety kept low (13) per xAI design philosophy. Cost/Speed penalized for 4-agent overhead.
← ALL HOSTS
COMPARE IN MATRIX →KNOWLEDGE GRAPH →